Systems and methods for water loss mitigation messaging

Abstract

A computer-implemented method, includes identifying, a set of insurance policyholders that have experienced water loss and a second set of insurance policyholders that have not experienced water loss. The method also includes determining an attribute indicative of increased likelihood of future water loss using a predictive model using a percentage of the first set of insurance policyholders defining a first sample size of the first set of insurance policyholders that is smaller relative to a percentage of the second set of insurance policyholders defining the second sample size of the second set of insurance policyholders. Further, the method includes identifying at least one targeted insurance policyholder having an increased likelihood of water loss, based upon the attribute and providing a water loss mitigation strategy to the at least one targeted insurance policyholder.

The invention claimed is: 
     
       1. A computer-implemented method, comprising:
 acquiring, from an analytical data store (ADS), policyholder data identifying an overall set of policyholders; 
 acquiring, from a claims analysis and reporting data store (CARDS), claims data indicative of properties that have incurred water loss; 
 generating model data, by merging the policyholder data with the claims data; 
 identifying, from the model data, a first set of policyholders that have experienced water loss and a second set of policyholders that have not experienced water loss, wherein the first set of policyholders is smaller than the second set of policyholders; 
 constructing a predictive water loss model that estimates a likelihood of a future water loss for the overall set of policyholders by data mining policy information associated with the overall set of policyholders to identify relationships between metrics of interest; 
 determining a size of a first sample of the first set of policyholders and a size of a second sample of the second set of policyholders such that the first sample of the first set of policyholders and the second sample of the second set of policyholders are balanced in a balanced data set, wherein the balanced data set is implemented by:
 identifying a first percentage of the overall set of policyholders that are the first set of policyholders; 
 defining the size of the first sample as an entirety of the first set of policyholders; 
 defining the size of the second sample as a second percentage of the second set of policyholders, wherein the second percentage is equal to the first percentage; and 
 creating the balanced data set by sampling the model data in accordance with the size of the first sample and the size of the second sample; 
 
 determining an attribute indicative of increased likelihood of future water loss via the predictive water loss model using the balanced data set; 
 based upon the attribute, identifying at least one targeted policyholder having an increased likelihood of water loss using a logistic regression that models a log-odds of water loss versus non-water-loss based on a linear combination of weighted predictor variables; and 
 providing a water loss mitigation strategy to the at least one targeted policyholder. 
 
     
     
       2. The computer-implemented method of  claim 1 , comprising identifying the first set of policyholders and the second set of policyholders by accessing and analyzing policy information stored in a database to distinguish the first set of policyholders from the second set of policyholders as those that have prior water loss claims, pending water loss claims, or both. 
     
     
       3. The computer-implemented method of  claim 1 , comprising providing the water loss mitigation strategy via an electronic message comprising a banner message of a policyholder portal, a hyperlink, an e-mail message, or any combination thereof. 
     
     
       4. The computer-implemented method of  claim 1 , comprising further defining the first sample by oversampling the first set of policyholders. 
     
     
       5. The computer-implemented method of  claim 1 , comprising further defining the second sample by randomly sampling the second set of policyholders. 
     
     
       6. The computer-implemented method of  claim 1 , comprising:
 determining predicted probabilities of water loss for each policyholder of the overall set of policyholders; and 
 identify the at least one targeted policyholder based upon the predicted probabilities. 
 
     
     
       7. The computer-implemented method of  claim 1 , comprising:
 identifying the at least one targeted policyholder based upon the predictive water loss model by segmenting the overall set of policyholders into a set of high risk policyholders and a set of low risk policyholders; and 
 identifying each of the set of high risk policyholders as the at least one targeted policyholder. 
 
     
     
       8. The computer-implemented method of  claim 7 , wherein the set of high risk policyholders are identified as those having a risk above a predetermined threshold value. 
     
     
       9. The computer-implemented method of  claim 7 , wherein the set of high risk policyholders are identified as those having a risk above an overall average of the overall set of policyholders. 
     
     
       10. A tangible, non-transitory, machine-readable medium, comprising machine-readable instructions that, when executed by one or more processors, cause the one or more processors to:
 acquire, from an analytical data store (ADS), policyholder data identifying an overall set of policyholders; 
 acquire, from a claims analysis and reporting data store (CARDS), claims data indicative of properties that have incurred water loss; 
 generate model data, by merging the policyholder data with the claims data; 
 identify, from the model data, a first set of policyholders that have experienced water loss and a second set of policyholders that have not experienced water loss, wherein the first set of policyholders is smaller than the second set of policyholders; 
 construct a predictive water loss model that estimates a likelihood of a future water loss for the overall set of policyholders by data mining policy information associated with the overall set of policyholders to identify relationships between metrics of interest; 
 determine a size of a first sample of the first set of policyholders and a size of a second sample of the second set of policyholders such that the first sample of the first set of policyholders and the second sample of the second set of policyholders are balanced in a balanced data set, wherein the balanced data set is implemented by:
 identifying a first percentage of the overall set of policyholders that are the first set of policyholders; 
 defining the size of the first sample as an entirety of the first set of policyholders; 
 defining the size of the second sample as a second percentage of the second set of policyholders, wherein the second percentage is equal to the first percentage; and 
 creating the balanced data set by sampling the model data in accordance with the size of the first sample and the size of the second sample; 
 
 determine an attribute indicative of increased likelihood of future water loss via the predictive water loss model using the balanced data set; 
 determine predicted probabilities of water loss for each policyholder of the overall set of policyholders; 
 based upon the attribute and the predicted probabilities, identify at least one targeted policyholder having an increased likelihood of water loss; and 
 provide a water loss mitigation strategy to the at least one targeted policyholder. 
 
     
     
       11. The tangible, non-transitory, machine-readable medium of  claim 10 , wherein the machine-readable instructions are configured to cause the one or more processors to provide the water loss mitigation strategy via an electronic message. 
     
     
       12. The tangible, non-transitory, machine-readable medium of  claim 10 , wherein the metrics of interest comprise a home age, a home size, a policy deductible amount, a policy loss surcharge indicator, a high-water-loss county indicator, a burglar alarm indicator, a number of policy non-payment cancellations, or a combination thereof. 
     
     
       13. The tangible, non-transitory, machine-readable medium of  claim 10 , wherein the machine-readable instructions are configured to cause the one or more processors to determine the attribute indicative of increased likelihood of future water loss based on data relating policy losses and policy coverage during a predetermined time period. 
     
     
       14. The tangible, non-transitory, machine-readable medium of  claim 10 , wherein the machine-readable instructions are configured to cause the one or more processors to identify the at least one targeted policyholder having an increased likelihood of water loss using a logistic regression that models a log-odds of water loss as opposed to non-water-loss based on a linear combination of weighted predictor variables. 
     
     
       15. A computer system, comprising:
 a memory; and 
 a processor configured to:
 acquire, from an analytical data store (ADS), policyholder data identifying an overall set of policyholders; 
 acquire, from a claims analysis and reporting data store (CARDS), claims data indicative of properties that have incurred water loss; 
 generate model data, by merging the policyholder data with the claims data; 
 identify, from the model data, a first set of policyholders that have experienced water loss and a second set of policyholders that have not experienced water loss by accessing and analyzing policy information stored in a database, wherein the first set of policyholders is smaller than the second set of policyholders; 
 construct a predictive water loss model that estimates a likelihood of a future water loss for the overall set of policyholders, by data mining policy information associated with the overall set of policyholders to identify relationships between metrics of interest; 
 determine a size of a first sample of the first set of policyholders and a size of a second sample of the second set of policyholders such that the first sample of the first set of policyholders and the second sample of the second set of policyholders are balanced in a balanced data set, despite the first set of policyholders being smaller than the second set of policyholders, by:
 identifying a first percentage of the overall set of policyholders that are the first set of policyholders; 
 defining the size of the first sample as an entirety of the first set of policyholders; 
 defining the size of the second sample as a second percentage of the second set of policyholders, wherein the second percentage is equal to the first percentage; and 
 creating the balanced data set by sampling the model data in accordance with the size of the first sample and the size of the second sample; 
 
 determine an attribute indicative of increased likelihood of future water loss via the predictive water loss model using the balanced data set; 
 determine predicted probabilities of water loss for each policyholder of the overall set of policyholders; 
 based upon the attribute and the predicted probabilities, identify at least one targeted policyholder having an increased likelihood of water loss; and 
 provide a water loss mitigation strategy to the at least one targeted policyholder. 
 
 
     
     
       16. The computing system of  claim 15 , wherein the processor is configured to identify the at least one targeted policyholder having an increased likelihood of water loss using a logistic regression that models a log-odds of water loss as opposed to non-water-loss based on a linear combination of weighted predictor variables. 
     
     
       17. The computing system of  claim 15 , wherein the processor is configured to:
 identify the at least one targeted policyholder based upon the predictive water loss model by identifying a set of high risk policyholders having a water loss risk that is above a predetermined threshold value or an overall average of the overall set of policyholders; and 
 identify each of the set of high risk policyholders as the at least one targeted policyholder. 
 
     
     
       18. The computing system of  claim 15 , wherein the database comprises an analytical data store (ADS), a claims analysis and reporting data store (CARDS), or both. 
     
     
       19. The computing system of  claim 15 , wherein the database comprises one or more tables implemented using DB2® relational database tables.